We present AnchorFlow-VIO, an open-source Basalt-derived monocular visual-inertial odometry system, and evaluate a uniform configuration on LaMAria. The system uses one grayscale camera and one IMU, combining factory IMU correction, a fixed timing adjustment, windowed initialization and fixed-template patch tracking. Server evaluation of all 63 test recordings yields control-point scores of 47.2, 37.1, 24.8, 35.1 and 15.6 for Short, Medium, Long, Low light and Moving platform. Test pose recalls at 5 m are 84.6%, 69.1%, 53.9% and 62.9% for the four categories reporting dense recall. On ten training recordings, mean control-point score is 41.38 and all 265,694 image timestamps receive poses, including 204 filled poses. One training platform recording diverges and scores zero. A paired four-record comparison favors fixed over rolling templates on three recordings, while reversing that ranking on the fourth. Increasing feature density also yields mixed results. Ten fresh runs at prospectively fixed 30 s and 120 s offsets expose substantial start sensitivity: a low-light recording scores 6.14 at the earlier start and 42.00 at the later start. These results characterize the limits of the frozen system and show why complete output, feature counts and selected successful runs are insufficient evidence of robust localization. Source code, frozen configuration and reproducible submission tools are publicly available; training ablations are reported separately from the submitted test evaluation.
Figures & tables
Test
CP score
R@5m (%)
Category
records
Training
Test
Training
Test
Short
18
74.05
47.2
100.00
84.6
Medium
10
33.68
37.1
70.48
69.1
Long
16
40.37
24.8
85.99
53.9
Low light
9
30.16
35.1
76.18
62.9
Moving platform
10
28.65
15.6
—
—
Table 1: Training and server-evaluated test results for the frozen method. Training has two recordings per category; test counts differ as shown. Different recordings prevent interpreting the differences as paired effects.
Figure 1: Training and test category scores and pose recall. Training means use two recordings each; test values are the official category display for the 18/10/16/9/10-record submission. Platform recall is unavailable in the summary.
Method
Short
Medium
Long
Low light
Platform
Mighty Camera
57.1 / 88.3
38.8 / 74.8
18.5 / 39.3
36.9 / 64.2
15.8
AnchorFlow-VIO
47.2 / 84.6
37.1 / 69.1
24.8 / 53.9
35.1 / 62.9
15.6
FPV-SLAM
47.0 / 81.0
28.4 / 67.0
25.7 / 54.6
45.0 / 80.7
22.5
ASTRM_FLOWSLAM_v1
46.9 / 76.2
30.1 / 56.7
26.0 / 58.1
34.3 / 59.3
23.5
AnonSLAM
43.3 / 91.9
33.6 / 78.9
30.0 / 76.9
41.7 / 88.3
19.1
(r) OpenVINS+Maplab
19.7 / 43.8
12.7 / 29.0
5.5 / 10.2
6.9 / 15.0
2.0
Table 2: Selected monocular + IMU entries from the official leaderboard on 19 October 2026. Cells show CP / R@5m (%); Platform shows CP only. Parenthetical “r” denotes organizer-reproduced runs. Values are server results, not reruns by the author, and no speed comparison is implied.
Record
CP
R@5m (%)
ATE (m)
Scale
Poses
Short19
79.02
100.00
0.59
0.9854
18,352
Short20
69.08
100.00
1.03
0.9789
20,262
Medium11
49.53
94.78
2.16
1.0051
23,748
Medium12
17.83
46.18
7.01
0.9508
33,513
Long17
48.90
92.97
2.44
0.9999
35,842
Long18
31.85
79.00
4.01
0.9850
33,878
Table 3: Frozen monocular policy on all ten challenge training recordings. R@5m is horizontal pose recall. ATE and scale use separate Sim(3) trajectory alignment. Dashes indicate unavailable dense reference metrics.
Figure 2: Estimated horizontal trajectories after subtracting each initial position. No ground truth, yaw alignment or scale fitting is applied. Axes are independent between panels. Platform12’s extreme extent exposes the failed estimate rather than a plausible traveled path. Green and orange markers show the first and last positions.
CP
R@5m (%)
Record
Fixed
Rolling
Fixed
Rolling
Short19
79.02
54.46
100.00
97.23
Medium12
17.83
33.26
46.18
78.64
Long17
48.90
17.13
92.97
35.68
Lowlight10
23.18
20.78
64.64
48.92
Table 4: Paired fixed versus rolling templates on four exploratory recordings. All other policy settings are unchanged.
Figure 3: Left: fixed-minus-rolling CP difference on the exploratory screen. Right: fitted similarity scale and aligned ATE for the eight full recordings with dense references. The dashed line marks unit scale.
Figure 4: Original and dense-feature CP scores on all ten recordings. The absent Lowlight11 dense bar is an alignment failure, not a measured zero. Zero scores on Platform12 are actual reported scores.
CP
R@5m (%)
Record
Start (s)
Fresh
Suffix
Fresh
Suffix
Short19
30
80.49
78.82
95.00
95.00
Medium12
30
15.55
16.72
43.72
43.26
Long17
30
45.40
46.21
90.50
90.94
Lowlight10
30
6.14
23.24
14.38
64.80
Platform12
30
2.85
0.00
—
—
Table 5: Fresh processes versus the original trajectory restricted to the same suffix. Both recalls retain the official full-reference denominator. No start-time winner is selected.
Figure 5: CP changes when starting fresh at 30 s and 120 s. Squares represent the original matched suffix; circles represent fresh runs. The large Lowlight10 change reverses direction across offsets.
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
Record
Original CP
Dense CP
Dense Δ CP
Short19
79.02
72.62
-6.40
Short20
69.08
58.81
-10.27
Medium11
49.53
42.46
-7.07
Medium12
17.83
48.98
+31.15
Long17
48.90
39.81
-9.09
Long18
31.85
20.32
-11.53
Appendix
Table 6: Complete feature-density comparison. The failed Lowlight11 alignment is retained as missing, while Platform12 contributes its measured zero.
Record / start
CPU (s)
Wall (s)
RSS (MiB)
Images/s
Short19 / 30 s
350.87
198.35
189.5
89.50
Medium12 / 30 s
628.27
378.82
270.9
86.88
Long17 / 30 s
676.92
395.03
288.6
89.21
Lowlight10 / 30 s
548.63
347.87
243.1
79.94
Platform12 / 30 s
498.20
294.12
246.5
88.62
Short19 / 120 s
317.35
179.78
180.9
88.73
Appendix
Table 7: Native controlled-start cost. RSS is the per-run process maximum, not aggregate service memory. Throughput excludes preparation and scoring.
Sequence
CP
CP R@1m (%)
Pose R@5m (%)
sequence_1_1
46.3
16.7
100.0
sequence_1_2
22.9
0.0
55.2
sequence_1_3
26.5
0.0
72.6
sequence_1_4
25.5
6.7
59.6
sequence_1_5
60.0
50.0
99.7
sequence_1_6
35.4
18.2
76.2
Appendix
Table 8: All 63 server-evaluated test recordings. Values reproduce the server display to one decimal; dashes are unavailable metrics. The single reported platform recall of zero is retained as displayed, not generalized to the other platform recordings.